Patterns of tobacco use and related factors among adolescents in South Africa : insight from the Global Youth Tobacco Survey

Abstract

INTRODUCTION : Adolescent multiple tobacco use is a major public health issue, both in South Africa and globally. This study sought to use latent class analysis to identify patterns of tobacco products use (combustible cigarettes, chewing tobacco, snuff) and combustible non-cigarette tobacco products smoking (such as cigars, little cigars, pipes) and related factors among adolescents in South Africa. METHODS : Data from the Global Youth Tobacco Survey (2011) were used. A nationally representative cross-sectional school-based survey was conducted among secondary school students using a stratified two-stage cluster sampling. The sample comprised 10822 students in grades 8 to 11 in South Africa (approximately aged 13–18 years). Latent class analysis was used to identify patterns of tobacco use and examined how sociodemographic and tobacco-related characteristics are associated with such patterns. Subsequently, the multinomial logistic regression analysis was used to assess the relationship between covariates and tobacco-related variables with the probability of belonging to a specific latent class. Results : The latent class analysis identified three classes: Class 1 (8.8%) was characterized by high probabilities of multiple tobacco product use; Class 2 (3.4%) had high probabilities of combustible non-cigarette tobacco products smoking; and Class 3 (83.9%) showed a minimal likelihood of current use across all four products. Compared to non-users, adolescents exposed to peer smoking had significantly higher odds of multiple product use (AOR=4.07; 95% CI: 2.93– 5.66) and combustible tobacco use (AOR=6.29; 95% CI: 4.15–9.53). Parental smoking was also associated with increased odds of both multiple (AOR=2.33; 95% CI: 1.81–3.00) and combustible (AOR=1.91; 95% CI: 1.31–2.79) tobacco use. Females had lower odds than males of using multiple (AOR=0.65; 95% CI: 0.49–0.87) and combustible products (AOR=0.51; 95% CI: 0.36–0.71). Older adolescents (aged ≥18 years) were more likely to be multiple users (AOR=4.18; 95% CI: 1.59–10.98). Support for smoke-free policies was associated with reduced odds of tobacco use, while knowledge of smoking harms was associated with combustible tobacco use (AOR=1.60; 95% CI: 1.07–2.39). CONCLUSIONS : Multiple tobacco use and combustible non-cigarette tobacco products smoking is common among South African adolescents. Understanding different patterns of multiple tobacco use and combustible non-cigarette tobacco products smoking can help to inform prevention and cessation programs for adolescents. Given the risk adolescents face, tobacco cessation interventions tailored to their tobacco product of choice are urgently needed.